Who
-
Participants: Hospitalized patients with tuberculosis (TB) in Japan
-
Sample size:
-
69,474 TB hospitalizations identified
-
19,930 patients met inclusion criteria
-
3,578 propensity score–matched pairs (7,156 patients)
-
-
Age: Mean age ≈ 80 years after matching; subgroup analyses for ≥75 vs <75 years
-
Key exclusions: Patients <15 years, pregnant patients, early deaths/discharges (≤2 days), and those not started on standard TB therapy promptly
What
-
Main finding: Adding pyrazinamide (PZA) to the initial TB regimen was not associated with increased in-hospital mortality in elderly patients.
-
Adverse events:
-
Overall adverse events were more frequent with PZA-containing regimens, driven mainly by allergic reactions.
-
No significant differences in hepatotoxicity, gout attacks, gastrointestinal intolerance, length of hospital stay, or 90-day readmission.
-
-
Conclusion: Advanced age alone should not be a reason to avoid PZA in TB treatment, though clinicians should monitor for allergic reactions.
When
-
Study period: July 2010 to March 2022
Where
-
Setting: Nationwide inpatient acute-care hospitals across Japan
-
Context: Japanese policy mandates hospitalization for sputum smear–positive TB until smear conversion
Why
-
Rationale:
-
PZA is often avoided in elderly TB patients due to safety concerns.
-
Existing evidence was limited to small studies with poor generalizability.
-
This study aimed to provide robust, nationwide evidence on the safety of PZA in older adults with TB.
-
How
-
Design: Retrospective nationwide cohort study
-
Data source: Japanese Diagnosis Procedure Combination inpatient database
-
Exposure groups:
-
HRE (isoniazid, rifampicin, ethambutol)
-
HREZ (isoniazid, rifampicin, ethambutol, pyrazinamide)
-
-
Analysis:
-
Propensity score matching to balance baseline characteristics
-
Primary outcomes: in-hospital mortality and composite adverse events
-
Secondary outcomes: length of stay, readmission, and medication use for complications
-
Sensitivity analyses stratified by age (≥75 vs <75 years)
No comments:
Post a Comment